In addition to efforts to describe the earliest signs of autism, it is critical to continue efforts to understand the brain-basis of autism, how it changes over time, and how it is related to the changing manifestations of the disorder in children and adults. The population of already-affected individuals with autism is huge and growing. It has become clear that autism in these individuals is a dynamic disorder and that a sizeable proportion of affected individuals worsen rather than improve as they grow older. Individuals whose impairment does not worsen over time live with stable but significant lifelong disability. The objective of this research is to understand longitudinal brain mechanisms from childhood into adulthood in autism and how dynamic brain and clinical phenotype changes are related. This research will result in new information important for the development of new treatments and secondary and tertiary preventive interventions. The specific aims of this project are to collect time 3 and time 4 longitudinal structural and diffusion tensor neuroimaging at 3 Telsa along with clinical and neuropsychological data on a large cohort of children and adults with autism and typically developing individuals. The collection of time 1 and time 2 data was funded by the NIH Collaborative Program of Excellence in Autism. This is a case-control longitudinal study of the original cohort of 100 males with autism and 72 matched normal controls. Subjects in the cohort are equally divided into 4 age-bins. At time 1 they were 3-6, 7-11, 12-17, and 18-35 years of age. At time 2 they were 2 years older. At times 3 and 4 they will be 4 and 6 years older. 3 Telsa structural and diffusion tensor imaging data were collected at time 1 and time 2 and will be collected at time 3 and 4 on the single, same scanner used for time 1 and 2. The 3 major and minor eigenvalues and eigenvectors are determined and fractional anisotropy, mean diffusivity, and axial and radial diffusivity calculated for quantitative assessment of white matter integrity. Region of interests and whole-brain voxel-based analyses of both structural and diffusion tensor imaging data are performed, along with tractography. Detailed clinical and neuropsychological data are collected on all subjects and dynamic brain-behavior relationships studied. This project is a collaborative effort between the University of Utah, where all subjects are ascertained, assessed, and scanned, the University of Wisconsin, where all diffusion tensor imaging data are processed and analyzed, and Brigham Young University where all structural data are processed and analyzed. This collaborative group has been effectively and productively working together on this longitudinal study since 2002.